[GRASSLIST:5658] Re: Creating DTM from shapefiles

[Michael Barton]

Maciek,

Not to worry. GRASS is a BIG program and hard to know what everything
does--especially across 2 versions. However, I've run into the same problem
recently and so have been testing various options of patching.

I *think* the script lets you patch points, lines, or areas (there is a
selection pull-down to choose). I also *think* it preserves the attribute
files. I don't know how it combines them however, especially if they have
different fields. It uses ogr2ogr, so the documentation to that may help--or
just a bit of trial and error.

I wrote it on a plane to a meeting in Washington, DC and so haven't had time
to test it much. I also need to add a brief routine at the end to clean up
the temporary files it makes and ask someone to test it for temp file
security (assuming I write the routine to delete the temporary files). But
it does show the advantages of using even simple shell scripting in GRASS.
The ease of making a nice GUI interface in scripts is thanks to Radim and
Markus. 

Michael
____________________
C. Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
PO Box 872402
Arizona State University
Tempe, AZ  85287-2402
USA

Phone: 480-965-6262
Fax: 480-965-7671
www: <www.public.asu.edu/~cmbarton>


> From: Maciek Sieczka <werchowyna at pf.pl>
> Date: Mon, 07 Feb 2005 00:32:33 +0100
> To: Michael Barton <michael.barton at asu.edu>, Maris Nartiss
> <maris.nartiss at gmail.com>, <grasslist at baylor.edu>
> Subject: Re: [GRASSLIST:5654] Re: Creating DTM from shapefiles
> 
> Guys
> 
> Michael is right. I'm sorry for my messing. Indeed only the cats are
> preserved in 5.4 v.patch. And indeed v.patch works the same in 5.7/6.0. When
> I wrote that labels are preserved by v.patch in 5.4 I was wrong. Patching
> cats is all I could do and why I came to conclussion that I managed to patch
> labels  once - I don't know. Obviously it mixed somehow in my mind and I
> wanted to be  the wise ass. That was dumb not to doublecheck what I was
> writing about before posting to the list. I hope I won't be taken as dumb in
> general, though :).
> 
> Thanks for the sript Michael. Haven't tried it yet. Does it let me to patch
> points and lines?
> 
> Maciek
> 
> P.S.
> Anyway, there could be some note in the v.patch manual regarding it's
> current limitations and eventual alternatives. Whom should I bother with
> this?
> 
> ----- Original Message -----
> From: "Michael Barton" <michael.barton at asu.edu>
> To: "Maciek Sieczka" <werchowyna at pf.pl>; "Maris Nartiss"
> <maris.nartiss at gmail.com>; <GRASSLIST at baylor.edu>
> Sent: Sunday, February 06, 2005 7:09 AM
> Subject: [GRASSLIST:5654] Re: Creating DTM from shapefiles
> 
> 
>> Maciek and Maris,
>> 
>> V.patch DOES work in GRASS 5.7/6.0. However, it works like it did in GRASS
>> 5.4. That is it simply combines vectors and their cat file. It does not
>> combine any attribute tables they may have.
>> 
>> I've written a simple script, using ogr2ogr, v.in.ogr, and v.out.ogr that
>> seems to do this. However, I haven't had much chance to test it yet. If
>> you'd like to give it a try, I attach it here.
>> 
>> Michael
>> ____________________
>> C. Michael Barton, Professor of Anthropology
>> School of Human Evolution and Social Change
>> PO Box 872402
>> Arizona State University
>> Tempe, AZ  85287-2402
>> USA
>> 
>> Phone: 480-965-6262
>> Fax: 480-965-7671
>> www: <www.public.asu.edu/~cmbarton>
>> 
>> 
>>> From: Maciek Sieczka <werchowyna at pf.pl>
>>> Date: Thu, 3 Feb 2005 19:07:16 +0100
>>> To: Maris Nartiss <maris.nartiss at gmail.com>, <GRASSLIST at baylor.edu>
>>> Subject: [GRASSLIST:5626] Re: Creating DTM from shapefiles
>>> 
>>> From: "Maris Nartiss" <maris.nartiss at gmail.com>
>>>> Hi,
>>>> 
>>>> I'm beginner with Grass and I have beginners problems :)
>>>> 
>>>> I have 2 shape files one with lines and second with points. Both of
>>>> them have HEIGHT parameter in attribute tables (lines - int, points -
>>>> float (double precision)). To do some modeling, I need to create
>>>> surface from those files.
>>>> I read some docs and understood, that best surface creation algorithm
>>>> in Grass it RST.
>>>> Now I have Q - how to combine both source vector data layers to feed
>>>> them to v.surf.rst?
>>>> I was trying 2 solutions, but, due to lack of knowledge, both of them
>>>> where dead ends.
>>>> 1) Using v.patch to combine both vector data sources
>>> 
>>> v.patch in 5.7/6.0 is not operational yet. You'll find some workarounds
>>> in
>>> recent archives of grass5 list. You can also try patching two such vector
>>> layers (points and lines) with v.patch in Grass 5.4/5.5. It should
>>> preserve
>>> labels of both input layers properly. But always check whether the labels
>>> were really preserved as they should. It happened to me with one
>>> particular
>>> dataset that they were not for some reason (though maybe it was the
>>> dataset
>>> fault). If all is ok then import into 6.0 with v.convert.
>>> 
>>>> 2) Convert lines to points and then combine them with v.patch
>>>> both worked fine but when I used d.what.vect I got only "Database
>>>> connection not defined".
>>> 
>>> v.patch doesn't output any database - so there is nothing to be
>>> connected.
>>> Moreover the vector layer resulting from v.patch in 5.7/6.0 has all the
>>> cats
>>> of the input layers preserved, so you result in having lots of features
>>> with
>>> the same cat, thus it is not virtually possible to prepare an appropriate
>>> database manually. All you get in an output of that v.patch is a
>>> vectorial
>>> drawing actually.
>>> 
>>>> Any ideas, links to howtos etc. welcome.
>>>> 
>>>> Maybe as a newbie with some ArcView background I'm unable to
>>>> understand some Grass specific things, who cause my problems, but
>>>> currently Grass docs aren't very helpful, mainly cuz they a bit
>>>> outdated :(
>>> 
>>> v.patch manual should contain some remarks on these issues until it's
>>> fixed
>>> IMHO.
>>> 
>>>> TNX for reading this,
>>>> Maris.
>>> 
>>> Cheers
>>> Maciek
>